Application Cluster Analysis on Time Series Modelling with Spatial Correlations for Rainfall Data in Jember Regency

Abstract

Forecasting is a statistical analysis to obtain an overview the development of event in the future. Forecasting performed on time series data, is a series of data observation data that affected by previous data. In addition, time series data is also affected by the location of research, it is called spatial correlations. This correlation can be analyzed by cluster analysis method. Cluster analysis aims to group objects based on similar characteristics. Variability of rainfall in Jember Regency depends on time and space so that there is a spatial correlation. Cluster analysis is expected to form groups that optimal in the data so that the forecasting results more optimal. Selection of the best forecasting models in this study is determined by the smallest RMSE value.